Ominous signs of a surge in new cases of drug-resistant HIV in newly infected, treatment-naïve patients have been documented in three recent studies, all of which report a heightened incidence of resistance to one or more classes of antiretroviral agents in the viral isolates of recently infected individuals. Although the actual numbers of patients with multi-drug-resistant HIV in these studies are small, the prevalence of this disturbing phenomenon is plainly on the increase.
These studies, all published within the last few months, involve patients who were infected during the year-long period when the data were being gathered. Researchers with the Aaron Diamond AIDS Research Center analyzed viral strains from 80 individuals who had been infected with HIV for an average of two months, and they found that 16% of these treatment-naïve individuals carried strains of HIV that were already resistant to at least one antiretroviral agent (1). Three of these patients harbored HIV strains that were considered multi-drug-resistant. The authors defined multi-drug resistance as combined genotypic and phenotypic resistance to two or more classes of antiretrovirals -- which is the generally accepted definition for a laboratory finding that has no strict definition at this time.
Little et al. found a similar rate of multi-drug resistance in newly infected individuals, suggesting that the incidence of primary infections that involve broadly resistant strains of HIV is currently between 2% and 4% in urban centers with high rates of HIV infection. This group of investigators studied 141 patients who had seroconverted within the previous 12 months. At the time resistance testing was done on viral isolates from these patients, none had been on antiretroviral therapy for more than seven days, yet three of these patients exhibited a greater than 10-fold reduction in susceptibility to one or more anti-HIV drugs, and two of the three had isolates resistant to two classes of drugs (2).
Isolates from 36 other subjects, 26% of the study population, showed lesser degrees of reduced susceptibility, with reductions ranging from 2.5- to 10-fold reduction.
Finally, Brodine et al. describe a cohort of 95 military personnel who were diagnosed with HIV infection in the period between February 1997 and February 1998. Thirty-one of these patients were treatment-naïve; of these, eight (26%) were infected with HIV strains that had resistance-conferring mutations in the reverse transcriptase or protease genes (3). One patient had mutations in both genes, suggesting that his viral strain would prove resistant to drugs that target either of these enzymes.
What do we do with this disturbing information? The rising number of cases of primary HIV infection that involve multi-drug-resistant viral strains has important implications both for the treatment of newly infected persons and for public-health prevention initiatives. It seems clear that if this trend continues -- and there is nothing to suggest that it won't -- care providers will have to give serious thought to whether they should order resistance testing for all newly diagnosed patients, to avoid starting those individuals on anti-retroviral regimens that are suboptimal.
Over the years clinicians who treat large numbers of HIV-infected patients have had to become amateur virologists, just to keep up with the latest developments in monitoring disease progression. For many, that education began with the advent of PCR-based viral load measurements (see "The HIV RNA Assay: A Valuable New Diagnostic Tool," Vol. 2, No. 2, and in that same issue, "Guidelines for the Clinical Use of the HIV RNA Assay." Both can be accessed at our new Web site, hivnewsline.com.) The next phase of that on-going education will involve learning when to order resistance assays -- and how to interpret the data derived from those assays.
The simplest of these resistance tests, the genotypic assay, detects point mutations in the viral genome that are known to be associated with resistance to specific antiretroviral agents. If, for example, a patient's viral isolate carries mutations at codon 82 of the protease gene, that patient is unlikely to respond to ritonavir or indinavir, but he or she may well prove responsive to saquinavir or nelfinavir (see "The Problem of Protease Resistance," Vol. 3, No. 3).
A potential advantage of genotypic analysis is that it permits the care provider to identify resistance-conferring mutations before they become clinically significant. A disadvantage is that because these assays are mutation-specific, they can tell us only what drugs are unlikely to benefit a given patient, not which drugs are most likely to confer a benefit. Moreover, these tests take time, and they are expensive. Genotypic assays are mere indicators, not predictors, of response -- and in heavily pretreated individuals the patient's history is likely to tell us as much about how he or she will respond to a new drug combination as a genotypic assay can tell us.
Several commercial laboratories now market genotype assays, but these tests are far from standardized. As the findings presented by Shuurman et al. at this fall's Interscience Conference on Antimicrobial Agents and Chemotherapy make abundantly clear, much work needs to be done to fine-tune this technology. Shuurman and coworkers sent identical viral samples to 33 different laboratories for genotyping. There was general agreement in results for samples with high-level resistance or no measurable resistance, but results were much less consistent for viral isolates that were known to exhibit moderate- or low-level resistance to a particular agent or class of agents (4).
The inescapable inference that one draws from the data collected by Shuurman et al. is that crucial decisions about the pharmacological management of patients with advanced HIV infection are being made based on the results of genotypic assays -- and, given the variability of the results obtained from different commercial laboratories, there is a real possibility that futile or even damaging decisions are sometimes being made.
Phenotypic assays provide clinicians with a considerably more accurate picture of how a particular patient will respond to particular antiretroviral agents -- but they do so at a considerably higher cost, one that can range from $600 to $900 for each agent tested. In a phenotype assay, viral strains are isolated from a patient's blood and mixed in the test tube with different concentrations of antiretroviral agents. The amounts of drug needed to inhibit the isolate's growth by 50% (IC50) and 90% (IC90) are obtained, thereby providing the clinician with a functional assessment of how susceptible a specific patient's viral isolate is to specific antiretroviral drugs. At this point only a relatively small number of commercial labs conduct phenotypic assays, but the results of these assays enjoy much greater reproducibility than genotype assays.
Several groups are developing consensus guidelines for the use of resistance testing in HIV infection, and the F.D.A. itself is studying the role of resistance testing in the treatment of seropositive patients. Until those guidelines are made available, care providers must use their own best judgment about when to order resistance tests.
Several companies have developed kits to test genotypic resistance, and if these kits gain F.D.A. approval, reimbursement ought to become easier to obtain. As with plasma HIV RNA testing, it is anticipated that the cost of resistance testing will decrease as such tests are incorporated into the routine management of HIV-infected individuals and the efficiencies of higher volume and competition drive down prices. Should newly diagnosed patients be tested for multi-drug-resistant HIV?
Many clinicians already use genotypic data to guide them in selecting a new combination of drugs for patients who are breaking through on their current regimen. An intriguing question raised by these studies is whether there is a role for resistance testing in treatment-naïve patients presenting with primary infection.
Pomerantz deems it reasonable to order resistance testing in all geographical locations where multi-drug-resistant strains of HIV are being reported among seroconverters (5). Other researchers disagree. They point out that, in the absence of drug therapy, drug-sensitive viral strains tend to multiply more rapidly than drug-resistant strains, which are less fit for viral replication (2, 6). For this reason, they contend, baseline genotypic results can be misleading. They give the care provider a susceptibility profile skewed in favor of the dominant strain of the virus, the strain that is most replication-competent. Under the pressure of multi-drug therapy, resistant viral strains may emerge, leading to rising viral burden and therapeutic failure.
It may be premature to make blanket recommendations about resistance testing in patients with primary HIV infection, but care providers who live in areas where cases of multi-drug-resistant primary infection are being reported may not want to wait for the publication of national guidelines before they begin to order resistance testing for recent seroconverters. We can only hope that the insurance industry will appreciate the cost-effectiveness of tailoring antiretroviral regimens a priori, rather than by trial and error. The once and future issue: Prevention.
When the data on multi-drug-resistant HIV are coupled with figures showing increased rates of transmission overall, the situation grows even more disheartening. This is especially true for people who have been living and working in the thick of this epidemic for the better part of two decades. Many of these individuals have buried spouses, partners, friends, siblings, parents, children. Some have seen their therapists and ministers, their bankers and bridge partners, their dentists and dog-walkers die of AIDS.
It is hardly surprising, then, that so many at-risk individuals are so depleted by the constant vigilance needed to remain HIV-negative. As Michael Shernoff, a New York City psychotherapist who treats many gay men, observes: "When I first began to develop AIDS-prevention interventions for gay men in the 1980s, I never dreamed that the lifestyle changes that we were encouraging people to make would have to be life long behavioral modifications. It is simply unrealistic to expect people to use a condom every time they have sexual relations, even when we know the risks involved in not using one. It is exhausting and counter-erotic to have to bring AIDS awareness to our love-making each and every time we get sexually involved with another person" (7).
Who can blame the communities that have been hardest hit by AIDS for embracing the notion that the end of the epidemic is in sight? What the rising rate of new HIV infections tells us is that the end is nowhere in sight. And what the escalating number of new infections involving broadly resistant strains of HIV tells us is that some individuals who know their HIV status -- and who have, moreover, been under treatment long enough to have developed resistance to more than one antiretroviral agent -- are not taking the precautions that need to be taken to avoid infecting others. Finally, what all of this tells us is that the front line of this epidemic is still buried in those secret, private places that physicians and social service providers are so powerless to penetrate.
Prevention workers face this challenge day in and day out. The motivations for engaging in risky sexual behavior are complex, says Bob Bergeron, the Director of Prevention Services at Gay Men's Health Crisis in New York. It is simplistic to attribute risky behavior to "battle fatigue" or to any other single cause, Bergeron says. The overall prevention message from GMHC has changed from a latex-centered one to a "harm reduction" model, one based on the belief that the most effective way to promote changes in sexual behavior is to determine the subtleties that cause people to take risks . . . and then to deal with those subtleties in specific, culturally sensitive ways (8).
Paradoxically, the successes achieved with combination therapy have made the work of prevention a harder sell. Most seropositive people don't look -- or feel -- sick anymore. Among young people, many of whom have yet to be disabused of the notion that they are immortal, there is a widespread belief that HIV infection is no longer a dire, life-threatening prospect. What they don't seem to have heard is that significant numbers of patients are experiencing disease progression on the potent new regimens now available to treat HIV infection. Do these young people know how toxic these medications are? Are they prepared to plan each and every day's activities around the inflexible demands of a multi-drug antiretroviral regimen? Perhaps the mainstream message of hope, aimed at people with HIV infection, was too successful. There was a healthy fear of HIV in the early 1980s, and that may not have been an altogether bad thing.
The three reports cited above are not the first to document the discovery of highly resistant HIV strains in treatment-naïve individuals; a group in San Francisco published a case report almost two years ago on a patient who had been infected with a viral strain so resistant to so many antiretroviral agents that he was essentially untreatable (9). But what this cluster of recent reports does suggest is that the prevalence of newly acquired multi-drug-resistant HIV is increasing sharply.
Before the incidence of new HIV infections involving multi-drug-resistant virus can double, and then double again, we need to formulate and implement public health initiatives that respond effectively to this considerable challenge. We need to alert the public, and government agencies, and at-risk individuals that a new wave of the epidemic is cresting -- before it bowls us over and leaves, in its wake, thousands of people who have acquired untreatable strains of the virus. If we are to curb the widespread transmission of untreatable strains of HIV, we need to reinvigorate, revivify, and reinvent our prevention efforts.